Instructions to use Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4 with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4

SGLang

How to use Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4 with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4 with Docker Model Runner:
```
docker model run hf.co/Baekpica/Solar-Open2-148B-A15B-REAP-184E-NVFP4
```

Baekpica commited on Jul 22

Commit

0508077

verified ·

1 Parent(s): a26c5c8

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -2

README.md CHANGED Viewed

@@ -84,7 +84,7 @@ The implementation follows and acknowledges the official [Cerebras REAP reposito
 ## Calibration Mixture
-REAP saliency used [Baekpica/K-EXAONE-236B-REAP-calibration-mix](https://huggingface.co/datasets/Baekpica/K-EXAONE-236B-REAP-calibration-mix) at pinned revision <code>94e90afc261634a30384d3037d9021f1c997b286</code>. The mixture contains **16,780 samples**, measured as exactly **106,795,574 tokens** by the pinned Solar tokenizer, with a maximum measured sample length of **269,944 tokens**. It was used for pruning saliency, not for activation calibration or fine-tuning.
 At layer 0, 23 routed experts received no route in this calibration mixture and consequently had exact zero observed saliency; all 23 were included among the 136 removed experts. Every routed expert in layers 1–47 was observed. The published calibration audit verifies the zero-valued accumulators, exact route totals, reconstructed saliency, and the final keep map rather than treating missing observations as measured evidence.
@@ -200,7 +200,7 @@ Solar Open 2는 fused expert tensor를 사용하므로 동일한 순서의 184
 ## Calibration 데이터 구성
-REAP saliency 계산에는 [Baekpica/K-EXAONE-236B-REAP-calibration-mix](https://huggingface.co/datasets/Baekpica/K-EXAONE-236B-REAP-calibration-mix)의 고정 revision <code>94e90afc261634a30384d3037d9021f1c997b286</code>을 사용했습니다. 전체 **16,780개 샘플**이며, 고정된 Solar tokenizer로 측정한 정확한 토큰 수는 **106,795,574개**, 측정된 샘플 최대 길이는 **269,944 tokens**입니다. 이 데이터는 pruning saliency에만 사용했고 activation calibration이나 fine-tuning에는 사용하지 않았습니다.
 0번 레이어에서는 routed expert 23개가 이 calibration mixture에서 한 번도 선택되지 않아 관측 saliency가 정확히 0이었고, 23개 모두 제거된 136개에 포함됐습니다. 1–47번 레이어에서는 모든 routed expert가 관측됐습니다. 공개 calibration audit은 미관측 항목의 accumulator가 정확히 0인지, 전체 route 수와 재구성한 saliency가 일치하는지, 최종 keep map에서 해당 expert들이 실제로 제거됐는지를 함께 검증합니다.

 ## Calibration Mixture
+REAP saliency used K-EXAONE-236B-REAP-calibration-mix that I used for the K-EXAONE-REAP series. The mixture contains **16,780 samples**, measured as exactly **106,795,574 tokens** by the pinned Solar tokenizer, with a maximum measured sample length of **269,944 tokens**. It was used for pruning saliency, not for activation calibration or fine-tuning.
 At layer 0, 23 routed experts received no route in this calibration mixture and consequently had exact zero observed saliency; all 23 were included among the 136 removed experts. Every routed expert in layers 1–47 was observed. The published calibration audit verifies the zero-valued accumulators, exact route totals, reconstructed saliency, and the final keep map rather than treating missing observations as measured evidence.
 ## Calibration 데이터 구성
+REAP saliency 계산에는 지난번 K-EXAONE-REAP 시리즈에 활용된 K-EXAONE-236B-REAP-calibration-mix를 사용했습니다. 전체 **16,780개 샘플**이며, 고정된 Solar tokenizer로 측정한 정확한 토큰 수는 **106,795,574개**, 측정된 샘플 최대 길이는 **269,944 tokens**입니다. 이 데이터는 pruning saliency에만 사용했고 activation calibration이나 fine-tuning에는 사용하지 않았습니다.
 0번 레이어에서는 routed expert 23개가 이 calibration mixture에서 한 번도 선택되지 않아 관측 saliency가 정확히 0이었고, 23개 모두 제거된 136개에 포함됐습니다. 1–47번 레이어에서는 모든 routed expert가 관측됐습니다. 공개 calibration audit은 미관측 항목의 accumulator가 정확히 0인지, 전체 route 수와 재구성한 saliency가 일치하는지, 최종 keep map에서 해당 expert들이 실제로 제거됐는지를 함께 검증합니다.